[1769] in linux-scsi channel archive
Re: SCSI Magneto-optical drive problem
daemon@ATHENA.MIT.EDU (Michael Weller)
Mon Apr 28 10:20:28 1997
Date: Mon, 28 Apr 1997 16:18:43 +0200 (MESZ)
From: Michael Weller <eowmob@exp-math.uni-essen.de>
Reply-To: Michael Weller <eowmob@exp-math.uni-essen.de>
To: "Dr H. T. Leung" <htl10@cus.cam.ac.uk>
Cc: linux-scsi@vger.rutgers.edu
In-Reply-To: <5jueom$a1@lyra.csx.cam.ac.uk>
On 27 Apr 1997, Dr H. T. Leung wrote:
> Sorry if this is an obvious question: I recently got hold of a magneto-optical
> drive, and all the SCSI-HOWTO, etc are fairly brief and just said they are used
> like floppies. Anyway, I did fdisk, then mke2fs, and mount. Then I can copy
> files, etc to the mounted disc. But then if I umount and change the disc and
> remount, it
Well, I can only guesse from your slightly foggy. However, this is what
can guess from my experience with just a similar M2512A and an AHA1742 (at
that time Adaptec seems to heave been moire friendly to the linux
community with regards to documentation and support).
In short: The Mo must detect a media change and signalit to the kernel
next time it is accessed and then the kernel must flush all buffers. It
might be a problem of your kernel or the somewhat bad reputationend 2940
driver, so you might try to upgrade to the most recent kernel you can get.
(recent ones even detect and properly support write protection (some older
ones did allow readwrite mountsbut created weirderrors when the Mo of
course refused to write to the MO).
Also, one can write tiny 10 line programs to enforce a buffer flush (I did
thatatmy firsttries with the MO, but I'm farfrom home right now and can't
access them this week).
However, I also found some funny thing about this MO: When jumpered for
Macintosh compatible operation it will NOT generate the required UNIT
attention state on meduia change. You must jumper it to not Mac compatible
for proper operation.
IMHO, all thisis in slight misinterpretation if not on contrary to the
SCSI 2 specs. but anyway. Rejumper and you should be fine.
Michael.
(eowmob@exp-math.uni-essen.de or eowmob@pollux.exp-math.uni-essen.de
Please do not use my vm or de0hrz1a accounts anymore. In case of real
problems reaching me try mat42b@spi.power.uni-essen.de instead.)
